필요 패키지¶

  • folium
  • lxml
  • geopandas 및 의존성 패키지들(https://thlee33.medium.com/windows-10%EC%97%90%EC%84%9C-geopandas-%EC%84%A4%EC%B9%98%ED%95%98%EA%B8%B0-2021-07-f07efea9294a)
In [ ]:
import folium

location = [37.2410864, 127.1775537]

def getNewMap():
  return folium.Map(location = location, tiles='cartodbpositron', zoom_start=11)

행정동 지도¶

In [ ]:
# geojson 형식의 파일을 dataframe으로 만들고 'adm_nm' 컬럼에 문자열 '용인시'를 포함하는 행만 추출

import geopandas as gpd

df = gpd.read_file('../data/dataset/행정동.geojson', encoding ='utf-8')
df = df[df['adm_nm'].str.contains('용인시')]
df.head()
Out[ ]:
OBJECTID adm_nm adm_cd adm_cd2 sgg sido sidonm sggnm adm_cd8 geometry
1464 1484 경기도 용인시처인구 포곡읍 3119111 4146125000 41461 41 경기도 용인시처인구 31191110 MULTIPOLYGON (((127.27860 37.30544, 127.28042 ...
1465 1485 경기도 용인시처인구 모현읍 3119112 4146125300 41461 41 경기도 용인시처인구 31191120 MULTIPOLYGON (((127.28334 37.34531, 127.28316 ...
1466 1486 경기도 용인시처인구 남사읍 3119114 4146125900 41461 41 경기도 용인시처인구 31191140 MULTIPOLYGON (((127.18870 37.08636, 127.18623 ...
1467 1487 경기도 용인시처인구 이동읍 3119113 4146125600 41461 41 경기도 용인시처인구 31191130 MULTIPOLYGON (((127.17441 37.22459, 127.17541 ...
1468 1488 경기도 용인시처인구 원삼면 3119134 4146134000 41461 41 경기도 용인시처인구 31191340 MULTIPOLYGON (((127.34840 37.20436, 127.34731 ...
In [ ]:
import matplotlib.pyplot as plt

df.plot(figsize=(6, 6))
plt.show()
In [ ]:
def draw_yongin_on_map(m):
  for _, r in df.iterrows():
    simplifed_geo = gpd.GeoSeries(r['geometry']).simplify(tolerance = 0.001)
    geo = simplifed_geo.to_json()
    geo = folium.GeoJson(data = geo, style_function= lambda x: {'fillColor' : 'orange', 'color': 'black'})
    folium.Popup(r['adm_nm']).add_to(geo)
    geo.add_to(m)

전기 충전소 지도¶

In [ ]:
import pandas as pd
# station_df = pd.read_xml('../data/dataset/전기충전소정보.xml', encoding='UTF-8')
station_df = pd.read_csv('../data/dataset/기존충전소위경도정보.csv', index_col = 0)
In [ ]:
station_df
Out[ ]:
이름 위도 경도
0 신동백 두산위브더제니스 37.306879 127.130787
1 죽전역 솔하임 37.328545 127.114735
2 광교상현 솔하임 37.302515 127.074668
3 성복역 롯데캐슬 클라시엘(파크나인2차) 37.319601 127.061733
4 까뮤이스테이트 37.335116 127.109361
... ... ... ...
1256 나인무인텔 37.208291 127.320402
1257 명가 37.277380 127.232383
1258 금화마을대우현대아파트 37.264701 127.110847
1259 플랫폼시티아이브오피스텔 37.298363 127.110283
1260 삼천리 에버랜드 복합충전소(완속) 37.293272 127.222609

1261 rows × 3 columns

In [ ]:
station_map = getNewMap()
draw_yongin_on_map(station_map)
station_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
positions = station_df[['위도', '경도']].values.tolist()

for position in positions:
      folium.Circle(position, radius= 100, color='#218721', fill=True).add_to(station_map)
      
station_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
station_map.save('../map/station.html')

수요 지도¶

TODO¶

  • resident, activity 구분한 수요 지도도 그리기
In [ ]:
import pandas as pd

resident_demand_df = pd.read_csv('../data/dataset/1.ev_app_resident.csv')
activity_demand_df = pd.read_csv('../data/dataset/2.ev_app_activity.csv')
total_demand_df = pd.read_csv('../data/dataset/4.ev_resident+ev_activity+ev_activity_resident.csv')
In [ ]:
# TODO 행정동 geo data에 합쳐서 지도에 수요 나타내기 -> Chropleth 맵 그리기
resident_demand_df = resident_demand_df[['adng_nm','cell_id', 'cell_xcrd', 'cell_ycrd', 'count_cust']]
activity_demand_df = activity_demand_df[['adng_nm','cell_id', 'cell_xcrd', 'cell_ycrd', 'count_cust']]
total_demand_df = total_demand_df[['adng_nm','cell_id', 'cell_xcrd', 'cell_ycrd', 'count_cust']]
In [ ]:
# 지역명으로 그룹화 후 수요를 합해 새로운 데이터프레임 생성

resident_demand_df = resident_demand_df.groupby('adng_nm')['count_cust'].sum().to_frame().reset_index()
activity_demand_df = activity_demand_df.groupby('adng_nm')['count_cust'].sum().to_frame().reset_index()
total_demand_df = total_demand_df.groupby('adng_nm')['count_cust'].sum().to_frame().reset_index()
In [ ]:
# 수요 데이터와 매핑하기 위해서 지리 데이터의 지역명 중 읍, 동만 남김

anm_nm = df['adm_nm'].apply(lambda x : x.split()[2])
df['adm_nm'] = anm_nm
In [ ]:
df.head()
Out[ ]:
OBJECTID adm_nm adm_cd adm_cd2 sgg sido sidonm sggnm adm_cd8 geometry
1464 1484 포곡읍 3119111 4146125000 41461 41 경기도 용인시처인구 31191110 MULTIPOLYGON (((127.27860 37.30544, 127.28042 ...
1465 1485 모현읍 3119112 4146125300 41461 41 경기도 용인시처인구 31191120 MULTIPOLYGON (((127.28334 37.34531, 127.28316 ...
1466 1486 남사읍 3119114 4146125900 41461 41 경기도 용인시처인구 31191140 MULTIPOLYGON (((127.18870 37.08636, 127.18623 ...
1467 1487 이동읍 3119113 4146125600 41461 41 경기도 용인시처인구 31191130 MULTIPOLYGON (((127.17441 37.22459, 127.17541 ...
1468 1488 원삼면 3119134 4146134000 41461 41 경기도 용인시처인구 31191340 MULTIPOLYGON (((127.34840 37.20436, 127.34731 ...
In [ ]:
def draw_demand_map(data_df):
  m = getNewMap()
  
  folium.Choropleth(
    geo_data = df,
    data = data_df,
    columns = ('adng_nm', 'count_cust'),
    key_on = "feature.properties.adm_nm",
    fill_color = "OrRd",
    legend_name = "지역별 수요",
  ).add_to(m)
  
  return m
In [ ]:
# 검은 색으로 나타나는 구역은 수요 정보가 없는 곳으로 추정

resdient_demand_map = draw_demand_map(resident_demand_df)
activity_demand_map = draw_demand_map(activity_demand_df)
total_demand_map = draw_demand_map(total_demand_df)
In [ ]:
resdient_demand_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
activity_demand_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
total_demand_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
result_df = pd.read_csv('../data/dataset/7.fin_location.csv', index_col=0)
In [ ]:
# 모든 후보군

candidate_map = getNewMap()
draw_yongin_on_map(candidate_map)

for i, r in result_df.iterrows():
  folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#0072B2', fill=True).add_to(candidate_map)

folium.LayerControl().add_to(candidate_map)
candidate_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
# 빨간색: 기존 충전소가 1km 반경 내에 설치되어 있는 후보군
# 파란색: 1km 반경 내에 충전소가 설치되어있지 않은 후보군

result_map = getNewMap()
draw_yongin_on_map(result_map)

for i, r in result_df.iterrows():
  # 1km 내에 존재하는 경우. 빨간색
  if r['existing']:
    folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#D95F02', fill=True).add_to(result_map)
  else:
    folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#0072B2', fill=True).add_to(result_map)
  
folium.LayerControl().add_to(result_map)
result_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
result_df = result_df[result_df['existing'] == 0]
result_map = getNewMap()
draw_yongin_on_map(result_map)

for i, r in result_df.iterrows():
    folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#0072B2').add_to(result_map)

folium.LayerControl().add_to(result_map)
result_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
result_df
Out[ ]:
cluster_이름 xcrds ycrds existing demand_satisfied
0 1001 127.295784 37.188830 0 0
1 1013 127.223315 37.135049 0 0
2 1014 127.213345 37.315307 0 0
3 1017 127.227875 37.354329 0 1
4 10206 127.419215 37.135348 0 0
... ... ... ... ... ...
903 963 127.368197 37.177893 0 1
904 974 127.367468 37.111187 0 0
905 976 127.222549 37.190262 0 0
906 99 127.252040 37.302776 0 1
907 997 127.317470 37.192700 0 1

908 rows × 5 columns

In [ ]:
result_map = getNewMap()
draw_yongin_on_map(result_map)

removed = 0

for i, r in result_df.iterrows():
  # 수요 15회를 만족하는 충전소
  if r['demand_satisfied']:
    folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#0072B2', fill=True).add_to(result_map)
  else:
    removed += 1
    folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#D63E2A', fill=True).add_to(result_map)
    

folium.LayerControl().add_to(result_map)
result_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
# 수요 62.5개 미만을 맡는 충전소
print(removed)

311
In [ ]:
result_df = result_df[result_df['demand_satisfied'] == 1]

result_map = getNewMap()
draw_yongin_on_map(result_map)

for i, r in result_df.iterrows():
    folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#0072B2', fill=True).add_to(result_map)
    

folium.LayerControl().add_to(result_map)
result_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [ ]:
from folium.plugins import BeautifyIcon

square_icon = BeautifyIcon(
    icon_shape='rectangle-dot', 
    border_color='red', 
    border_width=10,
)

result_map = getNewMap()
draw_yongin_on_map(result_map)

for i, r in result_df.iterrows():
  folium.Circle([r['ycrds'], r['xcrds']], radius= 100, color='#0072B2', fill=True).add_to(result_map)

for position in positions:
  folium.Circle(position, radius= 100, color='#218721', fill=True).add_to(result_map)
  
result_map
Out[ ]:
Make this Notebook Trusted to load map: File -> Trust Notebook